Data sharing

pdata_sharing <- df %>% 
  select(.run.number., .step., share.data.,
         mean.grants.groups:sum..total.datasets..of.groups) %>% 
  pivot_longer(-c(.run.number., .step., share.data.)) %>% 
  drop_na() %>% 
  mutate(share.data. = if_else(share.data., "Data sharing", "No data sharing"))

pdata_sharing %>% 
  filter(str_detect(name, "gini")) %>% 
  mutate(name = str_remove(name, "\\..*") %>% str_to_title()) %>% 
  ggplot(aes(.step., value, colour = share.data.)) +
  geom_smooth(se = FALSE) + # interval is anyways not showing up, so remove from legend
  facet_wrap(vars(name), nrow = 2, scales = "free_y") +
  scale_colour_viridis_d(option = "C", begin = .1, end = .9, alpha = .7) +
  labs(x = "step", y = "Gini index", colour = NULL) +
  theme(legend.position = c(.8, .5))

## `geom_smooth()` using method = 'gam' and formula 'y ~ s(x, bs = "cs")'

Initially higher inequality with data sharing, but equals out over time.

p <- pdata_sharing %>% 
  filter(str_detect(name, "gini")) %>% 
  ggplot(aes(.step., value, colour = share.data., group = .run.number.)) +
  geom_line() +
  facet_wrap(vars(name), nrow = 2)
plotly::ggplotly(p)

pdata_sharing %>% 
  filter(str_detect(name, "total\\.primary"),
         .step. == 500) %>% 
  ggplot(aes(share.data., value, fill = share.data.)) +
  geom_boxplot(notch = TRUE, show.legend = FALSE, width = .4, outlier.alpha = 0) +
  geom_jitter(alpha = .35, aes(fill = NULL), show.legend = FALSE, width = .1) +
  scale_fill_viridis_d(option = "C", begin = .1, end = .9, alpha = .5) +
  labs(x = NULL, y = "# of total publications produced") +
  coord_flip()